Twitter-based uncertainty and stock market returns: Evidence from G7 countries

Abstract

The aim of this study is to investigate the impact of Twitter-based economic uncertainty (TEU) and Twitter-based market uncertainty (TMU) on G7 stock returns in the challenging year in which the COVID-19 pandemic began (2020) under different stock market conditions (bearish, normal, and bullish). To this aim, this study applies novel quantile-based approaches, namely Quantile autoregression unit root test, Quantile-on-quantile approach, and Quantile Granger-causality test covering the period from 01 January 2020 to 15 September 2020. Main findings of the study are (1) G7 stock return series are stationary for all quantiles of the conditional distributions with minor exceptions meaning that shocks have temporary effects on stock returns of G7 markets. (2) The impact of Twitter-based uncertainty strongly depends on the market condition, whether it is bullish or bearish for all G7 markets. A heterogeneous association exists between variables caused by different market conditions. (3) A bi-directional causal association exists between stock returns-TEU and stock returns-TMU. This result confirms the existence of feedback hypothesis between G7 stock returns and TEU, TMU, respectively. This study provides important policy implications and recommendations for policy makers and investors on the nexus between Twitter-based uncertainties and stock returns.